Federated Learning and Generative AI for Resilient Supply Chain Network Optimization: A Data-Driven Approach
Abstract
Abstract The increasing complexity, interdependence, and volatility of global supply-chain networks require predictive models that can identify disruption risk while respecting organizational data boundaries. This paper develops an integrated Federated Learning–Generative Artificial Intelligence (FL–GenAI) framework in which supply-chain organizations collaboratively learn a disruption-prediction model without transferring raw operational records. The supply chain is represented as a directed graph whose nodes correspond to suppliers, manufacturing facilities, distribution centers, ports, transport fleets, and retail outlets, while edges represent material or information flows. At each edge node, a local generative model produces synthetic disruption observations to compensate for the scarcity of rare-event samples. The augmented local dataset is then used to train a predictive model, and only model updates are transmitted to a coordinating server. The framework explicitly addresses statistical heterogeneity, rare-event imbalance, communication cost, and privacy exposure. The simulated study described in the source manuscript contains 500,000 transit records distributed across 50 autonomous nodes and compares a centralized deep neural network, standard FedAvg, and the proposed FL–GenAI model. The reported results show accuracies of 95.2%, 86.4%, and 94.8%, respectively, while communication overhead is reported as 450 MB for centralized training and 85 MB for the federated…
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.